# CallMeGuLab/dsh-purge-safe > A DeepSeek Harness plugin that swaps system prompts and host policies on a local install for red-team evaluation and checking how tight the default restrictions are. - Magnitude: 5.7 out of 10 — Early signal - Stars: 515 total · +81 stars measured 2026-11-20, 00:09–03:39 UTC, ≈ 311 by evening - Star trust: star growth looks organic - Category: Security · Language: JavaScript · License: Apache-2.0 · Created: 2026-11-07 · Last push: 2026-11-19 - GitHub: https://github.com/CallMeGuLab/dsh-purge-safe · Page: https://gitnova.dev/en/r/CallMeGuLab/dsh-purge-safe ## Useful for - Check whether a host policy or the model itself blocks a response on your own local Harness install - Swap the default prompt for another rule set and revert it via Restore - Assess how strict the default permission policies and tool limits are in a controlled environment ## Why it’s here - Star-counter measurements on 2026-11-20 (UTC), 00:09–03:39: 434 → 515 stars (+81). This is the change over that interval. - Estimated end-of-day forecast: about +311 stars, using observed gains and the previous day. - Over the last two days the pace is 11× that of the previous week and a half. - The repository is 13 days old and already has 515 stars. With less than two weeks of history, there's no usual pace to compare the spike against yet. - About 280 forks a day — people are taking the code. ## Star trust Star growth looks organic. Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer. ## Numbers - Forks: 335 - Issues and pull requests: 1 - Watchers: 2 - Average over the last week: 93 per day - Usual pace: too little history (under two weeks) - Stars in the last hour (measured): 25 - Latest release: v2.0.0 (2026-11-19) ## Stars per day, last 14 days (oldest → newest, today is partial) 2026-11-07 … 2026-11-20: 10, 58, 33, 33, 22, 15, 8, 31, 22, 26, 11, 20, 229, 81 ## Magnitude by day, last 8 days 11-13 2.5, 11-14 2.7, 11-15 2.8, 11-16 2.3, 11-17 2.5, 11-18 2.0, 11-19 4.0, 11-20 5.7 ## More in this category 1. **BaaRai/ARTEX** — 8.6 · Breakout · Security · Go · +227 stars measured 2026-11-20, 00:08–03:39 UTC, ≈ 2,004 by evening Self-hosted LLM-driven penetration testing platform: a planner splits goals into intents, workers run real tools, and findings stay bound to HTTP traffic evidence. Full card: https://gitnova.dev/en/r/BaaRai/ARTEX.md 2. **elementalsouls/Claude-BugHunter** — 7.8 · Breakout · Security · Python · +227 stars measured 2026-11-20, 00:08–03:39 UTC, ≈ 1,224 by evening A Claude Code skill bundle that turns the assistant into a bug-hunting and external red-team helper: 83 skills, 15 slash commands and 681 patterns from disclosed reports across 24 vulnerability classes. Full card: https://gitnova.dev/en/r/elementalsouls/Claude-BugHunter.md 3. **MSNightmare/FastUnlock** — 7.2 · Early signal · Security · +41 stars measured 2026-11-20, 00:08–03:39 UTC, ≈ 360 by evening A 0day write-up of a Windows Narrator lockscreen bypass: with physical access it defeats BitLocker and achieves code execution as the WsiAccount user. Requires an online-linked account on the machine. Full card: https://gitnova.dev/en/r/MSNightmare/FastUnlock.md 4. **duwqijlk/google-trial-tool** — 6.8 · Early signal · Security · JavaScript · +40 stars measured 2026-11-20, 00:09–03:39 UTC, ≈ 285 by evening A JavaScript and Frida tool that injects into the Google One process to call the BillingClient API and activate hidden free Google One trial subscriptions. Runs in the MuMu emulator with a linked payment method. Full card: https://gitnova.dev/en/r/duwqijlk/google-trial-tool.md 5. **Hut-hzf/ung220z-tool** — 6.0 · Peaking · Security · Python · +47 stars measured 2026-11-20, 00:08–03:39 UTC, ≈ 298 by evening A tool for extracting credentials from the UNG220Z optical modem (Unionman/China Mobile): it reads and decrypts the device configuration over Telnet, printing current Web, Telnet, PPPoE, WiFi and root passwords. Full card: https://gitnova.dev/en/r/Hut-hzf/ung220z-tool.md --- Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score. Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast. Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself. Data as of 2026-11-20 03:48 UTC, updated every 30 minutes.